This proposal addresses (I) the epigenetic control of cancer and (II) the relation between stem cells and cancer stem cells. I. Genome-wide hypomethylation, leading to genomic instability and increased rates of LOH, and regional hypermethylation, leading to silencing of tumor suppressor genes, are both hallmarks of cancer. The focus of Aim 1 is to establish a causal relation between genomic methylation and tumor formation in multistage cancers of the intestine, pancreas and prostate. Two complementary approaches will be pursued that either cause inhibition or ectopic activation of Dnmtl and Dnmt3b, two methyltransferases (MTases) that are crucial for maintaining and establishing genomic methylation. Because epigenetic changes - in contrast to genetic alterations - are reversible, the DNA methyltransferases represent promising targets for drug therapy. This proposal seeks to define the role these enzymes play in the development of cancers in * different tissues. II. Most cancers comprise a heterogeneous population of cells with marked differences in their proliferative potential. In the cancer stem cell concept, minor populations of tumor cells that possess the stem cell property of self-renewal are the only cells that are immortal and can sustain tumor growth. It is the second focus of this proposal to define epigenetic and genetic differences that distinguish normal stem cells from cancer stem cells. Because cancer stem cells are thought to be resistant to conventional cancer therapy it is vital to understand the differences between the tumor-initiating stem cells and the bulk tumor cells on a molecular level. One of the key issues for our understanding of the relation between adult stem cells and cancer stem cells is defining the molecular circuitry that drives self renewal in the normal stem cell and its malignant counterpart. We have shown that activation of Oct4 in cells of the adult mouse leads to a rapid and fully reversible dysplastic proliferation of the intestinal and skin stem/progenitor cell compartment. Oct4 is a pluripotency gene that in combination with Nanog and Sox2 controls the self-renewal of embryonic stem cells. In Aim 2 we will investigate whether these key genes of ES cell renewal function also in self-renewal of adult stem cells and assess their possible role in cancer. We have used nuclear cloning as a novel and unbiased experimental tool to reprogram the genome of cancer cells and to distinguish between genetic and epigenetic changes that determine the malignant cell phenotype. Aim 3 will use nuclear transplantation as a criterion to define the genetic and epigenetic alterations of the cancer cell genome that interfere with reversion to pluripotency.